app.py

# app.py
import os
import gradio as gr
from transformers import pipeline
from diffusers import StableDiffusionPipeline

# Install necessary dependencies
os.system("pip install transformers diffusers gradio")

# Text Understanding with Hugging Face Transformers
def analyze_text(input_text):
    summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
    summary = summarizer(input_text, max_length=50, min_length=10, do_sample=False)[0]['summary_text']

    ner_pipeline = pipeline("ner", model="dbmdz/bert-large-cased-finetuned-conll03-english")
    entities = ner_pipeline(input_text)
    key_entities = [entity["word"] for entity in entities]

    return summary, key_entities

# Generate Images Using Stable Diffusion
def generate_images(prompts):
    model_id = "CompVis/stable-diffusion-v1-4"
    sd_pipeline = StableDiffusionPipeline.from_pretrained(model_id)
    images = []
    for prompt in prompts:
        image = sd_pipeline(prompt).images[0]
        images.append(image)
    return images

# Gradio App Interface
def generate_video_from_text(input_text):
    # Analyze text
    summary, key_entities = analyze_text(input_text)
    
    # Generate prompts and images
    prompts = [f"{entity}, cinematic, ultra-realistic" for entity in key_entities]
    images = generate_images(prompts)

    return summary, images

interface = gr.Interface(
    fn=generate_video_from_text,
    inputs="text",
    outputs=["text", gr.outputs.Image(type="pil", label="Generated Images")],
    title="Hugging Face Text Analysis & Image Generator",
    description="Analyze text to extract key entities and generate corresponding images using open-source AI models."
)

if __name__ == "__main__":
    interface.launch()